The goal of this project is to use functional magnetic resonance imaging (FMRI) and arterial spin labeling (ASL) to examine how cardiovascular disease (CVD) affects cognitive function and the measurement of brain function. CVD is associated with cognitive deficits; however, it is not understood how abnormal cardiac output associated with severe CVD leads to impaired cognitive function. There is increasing evidence that patients with severe CVD experience diminished cerebrovascular function, which ultimately affects cognitive performance. FMRI and ASL provide reliable methods to examine the intervening links between systemic vascular dysfunction and cognitive performance. The blood oxygen level dependent (BOLD) FMRI response itself has been assumed to be constant across patient populations in previous research. Yet,there is evidence that the integrity of the BOLD response changes with vascular integrity. Studies have found that patients with CVD exhibit changes in the BOLD response; however, it remains to be determined if this observation is due to neural function, cerebrovascular perfusion, or both. There are two likely mechanisms to explain the relationship between systemic vascular dysfunction and observed reductions on measures of cognitive function: 1) chronic cerebrovascular dysfunction (e.g., hypoperfusion) secondary to severe CVD might cause neural changes and subsequently diminished cognitive function; 2) since BOLD FMRI assesses the delivery of oxygen-rich blood in excess of demand, acute cerebral hypoperfusion might also directly attenuate the hemodynamic response measured by FMRI, independent of neuronal function. This potential decoupling of neural demands and observed BOLD response needs to be examined to address validity and reliability of FMRI among this population. We will directly examine these possibilities by assessing systemic cardiovascular function, cerebrovascular hemodynamic functions, and cognitive performance among 60 elderly patients with severe CVD and 30 age-matched controls. Cognitive measures were chosen because they have elicited performance and FMRI abnormalities among CVD patients in the past. Control paradigms will be administered to assess BOLD response itself. Groups will be contrasted on BOLD response to three levels of cognitive challenge: none (i.e., hypercapnia), matched (sensory and motor), and difficult cognitive paradigms. Further analyses will be conducted among the CVD group to quantify the contributions of blood perfusion. We predict that BOLD response will be diminished during all conditions among the CVD group, and that a significant portion of this reduction will be attributed vascular hypoperfusion. With rapidly growing clinical and experimental applications, it is important to demonstrate reliability and validity of FMRI in CVD. Findings are expected to provide data and methods that will increase the clinical utility of FMRI for the assessment of CVD patients, and a better understanding of the mechanisms causing cognitive impairment.